Concrete actions to improve participation and representation in Materials Project open-source

[mkhorton]

I'm going to add some suggestions myself as a current community member, but please add your own suggestions too, so that there can be a robust discussion at the next Foundation meeting.

Some ideas:

1. Formal or informal surveys to talk to people who may want to contribute and have not. This could be done at a large materials science conference, and could include co-ordination with the conference organizers for a more holistic discussion on open-source participation.
2. Encourage community leaders to attend courses provided by the Center for Scientific Collaboration and Community Engagement.
3. The GitHub Actions for statistics on issues, pull requests and discussions could be installed in all repos. Time to first response is a commonly-quoted metric for determining if a contributor is likely to continuing staying involved.
4. Provide additional training and documentation for new contributors.
5. [Your idea here!]

[rkingsbury]

I imagine many new contributors (esp. those from groups underrepresented in programming or MatSci) are quite intimidated about opening public issues or PRs, out of concern for appearing ignorant or being judged or criticized.

To help address that, I wonder if we could have some sort of process whereby a new contributor could request a private review of their work, possibly accompanied by a brief "live" 1:1 meeting to give feedback?

[mkhorton]

I wonder if we could have some sort of process whereby a new contributor could request a private review of their work, possibly accompanied by a brief "live" 1:1 meeting to give feedback?

I like this idea. A half-hour "submit your first PR together" coaching session or similar?

[mkhorton]

Re. suggestion 3, I added the GitHub action for metrics to the crystal toolkit and pymatgen repos as a trial. This will run once a month so we may need some time to build up data.

[JaGeo]

Could we organize an online event similar to the LLM hackathons to raise interest into contributing to our codes? If we would do this in advance of a new pymatgen paper, we could then also easily add new major contributors to the author list. It would mean some organization effort from our side. However, it might be paying off.

[janosh]

If we would do this in advance of a new pymatgen paper, we could then also easily add new major contributors to the author list.

i certainly like the idea but don't have the capacity to help organize this. i also wonder how much participation we would get given long-term OSS projects aren't as hot a topic as LLMs.

[mkhorton]

Re. 4, we discussed creating a sign-up sheet for mentorship from more senior contributors, for example people willing to donate 1hr/month to open source mentorship, as a concrete next step.

[JaGeo]

Instead of signing up for an hour per months, people might also volunteer to make a video on how to submit a PR to one of our repos (or we link to good material from other sources)

[JaGeo]

I have added a google doc to our slack channel to sign up as a mentor. Next time, we need to discuss about how we advertise this and how to run this project (especially, how people can contact us: mail address, forum?).

[JaGeo]

Potential Actions:

1. Pin a specific issue with info on the mentorship programs in the issues of each repo or in the readme files

@utf would it be okay if we pin such an issue to the atomate2 repo?
@mkhorton @janosh would it be okay if we pin such an issue to the pymatgen repo?

2. Advertise Contributions/Mentorship option to potential PIs - we would need to come up with an email draft and who to approach.

3. Add additional members to MP foundation (current attendances are quite low)

[rkingsbury]

Suggested decision outcome:

• first PR is often obstacle to growing developer base
• For first time contributors, working on issues flagged a certain way, we'll provide 1:1 mentoring to get the PR merged
• Eventually will provide networking and prof dev opportunity for postdocs / grad students

[rkingsbury]

I've updated the PR text with a more elaborated form of the proposal we discussed today. We will vote on this at the next meeting.